v16i064: nlmdl - Estimate nonlinear statistical models, Part02/06

Ron Gallant arg at ccvr1.cc.ncsu.edu
Sat Jan 12 16:05:46 AEST 1991


Submitted-by: arg at ccvr1.cc.ncsu.edu (Ron Gallant)
Posting-number: Volume 16, Issue 64
Archive-name: nlmdl/part02

# This is a shell archive.  Remove anything before this line, then unpack
# it by saving it into a file and typing "sh file".  To overwrite existing
# files, type "sh file -c".  You can also feed this as standard input via
# unshar, or by typing "sh <file", e.g..  If this archive is complete, you
# will see the following message at the end:
#		"End of archive 2 (of 6)."
# Contents:  ch5eg1/detail.bak ch5eg1/end3l.bak ch5eg1/ending.bak
#   ch6eg1/ending.bak ch6eg2/detail3l.bak display.cc nlmdl.h realmat.h
# Wrapped by arg at sparc on Fri Jan 11 20:14:53 1991
PATH=/bin:/usr/bin:/usr/ucb ; export PATH
if test -f 'ch5eg1/detail.bak' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'ch5eg1/detail.bak'\"
else
echo shar: Extracting \"'ch5eg1/detail.bak'\" \(5719 characters\)
sed "s/^X//" >'ch5eg1/detail.bak' <<'END_OF_FILE'
XMissing observation at t = 142.
XMissing observation at t = 170.
XMissing observation at t = 172.
XMissing observation at t = 193.
XWarning, status::to, var is not initialized, tmp.dat incomplete.
X
X
X     **********************************************************************
X     *                                                                    *
X     *                             nlmdl 2.0                              *
X     *                                                                    *
X     **********************************************************************
X
X
X     **********************************************************************
X     *                                                                    *
X     *                         Parameter settings                         *
X     *                                                                    *
X     **********************************************************************
X
X     Gallant, "Nonlinear Statistical Models," Chapter 5, Figure 3, p. 324-326
X     SUR                What estimation method?  Code SUR, TSLS, or GMM.
X     224                Number of observations, t = 1, ..., n.
X     2                  Number of equations, i.e. dimension of e.
X     0                  Number of instruments, i.e. dimension of Z.
X     8                  Number of parameters, i.e. dimension of theta.
X     50                 Upper limit on Gauss-Newton iterations.
X     1                  Number var iterates, ivar=0 means none.
X     homoskedastic      Code homoskedastic or heteroskedastic.
X     0                  Number of moving average terms MA for var estimate.
X     none               Code none or Parzen, none when MA>0 is unwise.
X     1.000000e-13       Convergence tolerance, tol=1.0e-8 is reasonable.
X     1.000000e-10       Inversion tolerance, eps=1.0e-13 is reasonable
X     minimal            How much output?  Code none, minimal, or full.
X
X
X     **********************************************************************
X     *                                                                    *
X     *                           Starting theta                           *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                           Col  1
X 
X                               Row  1    -2.90000
X                               Row  2    -1.30000
X                               Row  3    0.820000
X                               Row  4    0.360000
X                               Row  5    -1.50000
X                               Row  6    -1.00000
X                               Row  7  -0.0300000
X                               Row  8   -0.470000
X
X
X     **********************************************************************
X     *                                                                    *
X     *                               theta                                *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                           Col  1
X 
X                               Row  1    -2.92458
X                               Row  2    -1.28675
X                               Row  3    0.818570
X                               Row  4    0.361158
X                               Row  5    -1.53759
X                               Row  6    -1.04896
X                               Row  7   0.0300867
X                               Row  8   -0.467420
X
X
X
X     **********************************************************************
X     *                                                                    *
X     *                                var                                 *
X     *                (no degrees of freedom corrections)                 *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                     Col  1      Col  2
X 
X                         Row  1    0.164925   0.0920057
X                         Row  2   0.0920057   0.0896426
X
X
X
X     **********************************************************************
X     *                                                                    *
X     *                                 V                                  *
X     *                             (rank = 8)                             *
X     *                                                                    *
X     **********************************************************************
X 
X 
X             Col  1      Col  2      Col  3      Col  4      Col  5      Col  6
X 
X Row  1   0.0760485  -0.0394416 -0.00813632  0.00225515   0.0151611 -0.00086387
X Row  2  -0.0394416   0.0506098  0.00219396  0.00071347 -0.00244522   0.0122632
X Row  3 -0.00813632  0.00219396  0.00644155  0.00078138 -0.00603532 -0.00238599
X Row  4  0.00225515  0.00071347  0.00078138  0.00090344 -0.00048087  0.00101291
X Row  5   0.0151611 -0.00244522 -0.00603532 -0.00048087  0.00832137  0.00200847
X Row  6 -0.00086387   0.0122632 -0.00238599  0.00101291  0.00200847  0.00689444
X Row  7 -0.00293607  0.00321764 -0.00075345 -0.00079074  0.00110681  0.00015596
X Row  8 -0.00218205  0.00172663  0.00080830  0.00027549 -0.00115374  0.00049917
X 
X 
X             Col  7      Col  8
X 
X Row  1 -0.00293607 -0.00218205
X Row  2  0.00321764  0.00172663
X Row  3 -0.00075345  0.00080830
X Row  4 -0.00079074  0.00027549
X Row  5  0.00110681 -0.00115374
X Row  6  0.00015596  0.00049917
X Row  7  0.00128616 -0.00024276
X Row  8 -0.00024276  0.00036532
X
END_OF_FILE
if test 5719 -ne `wc -c <'ch5eg1/detail.bak'`; then
    echo shar: \"'ch5eg1/detail.bak'\" unpacked with wrong size!
fi
# end of 'ch5eg1/detail.bak'
fi
if test -f 'ch5eg1/end3l.bak' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'ch5eg1/end3l.bak'\"
else
echo shar: Extracting \"'ch5eg1/end3l.bak'\" \(5775 characters\)
sed "s/^X//" >'ch5eg1/end3l.bak' <<'END_OF_FILE'
XGallant, "Nonlinear Statistical Models," Chapter 5, Figure 3, p. 324-326
XSUR                What estimation method?  Code SUR, TSLS, or GMM.
X224                Number of observations, t = 1, ..., n.
X2                  Number of equations, i.e. dimension of e.
X0                  Number of instruments, i.e. dimension of Z.
X8                  Number of parameters, i.e. dimension of theta.
X50                 Upper limit on Gauss-Newton iterations.
X1                  Number var iterates, ivar=0 means none.
Xheteroskedastic    Code homoskedastic or heteroskedastic.
X3                  Number of moving average terms MA for var estimate.
XParzen             Code none or Parzen, none when MA>0 is unwise.
X1.000000e-13       Convergence tolerance, tol=1.0e-8 is reasonable.
X1.000000e-10       Inversion tolerance, eps=1.0e-13 is reasonable
Xfull               How much output?  Code none, minimal, or full.
X8                  Computed rank of V.
Xuncorrected        Divisor of var, corrected or uncorrected.
X                   Blank line.
X                   Blank line.
X                   Blank line.
X  -2.9245814501616803e+00          -2.92458145  theta(1)
X  -1.2867460434063589e+00          -1.28674604  theta(2)
X   8.1856990357452608e-01           0.81856990  theta(3)
X   3.6115785200404138e-01           0.36115785  theta(4)
X  -1.5375885808695555e+00          -1.53758858  theta(5)
X  -1.0489590970263374e+00          -1.04895910  theta(6)
X   3.0086700241162653e-02           0.03008670  theta(7)
X  -4.6742011797582117e-01          -0.46742012  theta(8)
X   1.6492462994112111e-01           0.16492463  var(1,1)
X   9.2005729031321579e-02           0.09200573  var(2,1)
X   9.2005729031321579e-02           0.09200573  var(1,2)
X   8.9642642316875812e-02           0.08964264  var(2,2)
X   6.4151871786190293e-02           0.06415187  V(1,1)
X  -2.6808435093990508e-02          -0.02680844  V(2,1)
X  -1.0481302083617307e-02          -0.01048130  V(3,1)
X   2.4179792658493493e-03           0.00241798  V(4,1)
X   1.5271744257123141e-02           0.01527174  V(5,1)
X   3.6217216495842341e-03           0.00362172  V(6,1)
X  -2.2745052058113957e-03          -0.00227451  V(7,1)
X  -2.2282353784641093e-03          -0.00222824  V(8,1)
X  -2.6808435093990942e-02          -0.02680844  V(1,2)
X   3.9204445770723149e-02           0.03920445  V(2,2)
X   3.5239211150354388e-03           0.00352392  V(3,2)
X   1.9044116642894958e-03           0.00190441  V(4,2)
X  -2.5090660931227499e-03          -0.00250907  V(5,2)
X   9.9860387081928022e-03           0.00998604  V(6,2)
X   1.0763522857786497e-03           0.00107635  V(7,2)
X   1.4608657346846976e-03           0.00146087  V(8,2)
X  -1.0481302083617045e-02          -0.01048130  V(1,3)
X   3.5239211150353313e-03           0.00352392  V(2,3)
X   5.9499339084749959e-03           0.00594993  V(3,3)
X   3.8379282539160748e-04           0.00038379  V(4,3)
X  -5.5488697133901426e-03          -0.00554887  V(5,3)
X  -2.3181283618853221e-03          -0.00231813  V(6,3)
X  -3.0995472128008387e-04          -0.00030995  V(7,3)
X   7.9619528555448433e-04           0.00079620  V(8,3)
X   2.4179792658493345e-03           0.00241798  V(1,4)
X   1.9044116642895203e-03           0.00190441  V(2,4)
X   3.8379282539158981e-04           0.00038379  V(3,4)
X   1.1185997609060533e-03           0.00111860  V(4,4)
X  -8.2948235409772357e-05          -0.00008295  V(5,4)
X   1.8128096548391446e-03           0.00181281  V(6,4)
X  -8.6907934492451726e-04          -0.00086908  V(7,4)
X   2.3704685064746336e-04           0.00023705  V(8,4)
X   1.5271744257122990e-02           0.01527174  V(1,5)
X  -2.5090660931226623e-03          -0.00250907  V(2,5)
X  -5.5488697133901886e-03          -0.00554887  V(3,5)
X  -8.2948235409779458e-05          -0.00008295  V(4,5)
X   7.2303542854823596e-03           0.00723035  V(5,5)
X   2.4292056516279819e-03           0.00242921  V(6,5)
X   6.6272949286796750e-04           0.00066273  V(7,5)
X  -1.0628753376875112e-03          -0.00106288  V(8,5)
X   3.6217216495840446e-03           0.00362172  V(1,6)
X   9.9860387081929288e-03           0.00998604  V(2,6)
X  -2.3181283618853659e-03          -0.00231813  V(3,6)
X   1.8128096548391255e-03           0.00181281  V(4,6)
X   2.4292056516280035e-03           0.00242921  V(5,6)
X   6.9725212588786700e-03           0.00697252  V(6,6)
X  -8.0090995756335992e-04          -0.00080091  V(7,6)
X   2.8957023926462434e-04           0.00028957  V(8,6)
X  -2.2745052058114465e-03          -0.00227451  V(1,7)
X   1.0763522857786777e-03           0.00107635  V(2,7)
X  -3.0995472128007016e-04          -0.00030995  V(3,7)
X  -8.6907934492452008e-04          -0.00086908  V(4,7)
X   6.6272949286796273e-04           0.00066273  V(5,7)
X  -8.0090995756337260e-04          -0.00080091  V(6,7)
X   1.1126181202496821e-03           0.00111262  V(7,7)
X  -2.5799316642020757e-04          -0.00025799  V(8,7)
X  -2.2282353784641279e-03          -0.00222824  V(1,8)
X   1.4608657346847011e-03           0.00146087  V(2,8)
X   7.9619528555449908e-04           0.00079620  V(3,8)
X   2.3704685064746418e-04           0.00023705  V(4,8)
X  -1.0628753376875236e-03          -0.00106288  V(5,8)
X   2.8957023926462423e-04           0.00028957  V(6,8)
X  -2.5799316642020887e-04          -0.00025799  V(7,8)
X   3.6558708385757817e-04           0.00036559  V(8,8)
X   7.1212458944209244e-09           0.00000001  D(1)
X  -9.5291010557382897e-09          -0.00000001  D(2)
X  -1.5778368693623276e-09          -0.00000000  D(3)
X  -6.4433263552306315e-10          -0.00000000  D(4)
X   1.5722264246975941e-09           0.00000000  D(5)
X  -2.4312439036197547e-09          -0.00000000  D(6)
X  -1.5845329475213192e-10          -0.00000000  D(7)
X  -8.7707819833213637e-10          -0.00000000  D(8)
X   4.4685695288184939e+02         446.85695288  obj
END_OF_FILE
if test 5775 -ne `wc -c <'ch5eg1/end3l.bak'`; then
    echo shar: \"'ch5eg1/end3l.bak'\" unpacked with wrong size!
fi
# end of 'ch5eg1/end3l.bak'
fi
if test -f 'ch5eg1/ending.bak' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'ch5eg1/ending.bak'\"
else
echo shar: Extracting \"'ch5eg1/ending.bak'\" \(5775 characters\)
sed "s/^X//" >'ch5eg1/ending.bak' <<'END_OF_FILE'
XGallant, "Nonlinear Statistical Models," Chapter 5, Figure 3, p. 324-326
XSUR                What estimation method?  Code SUR, TSLS, or GMM.
X224                Number of observations, t = 1, ..., n.
X2                  Number of equations, i.e. dimension of e.
X0                  Number of instruments, i.e. dimension of Z.
X8                  Number of parameters, i.e. dimension of theta.
X50                 Upper limit on Gauss-Newton iterations.
X1                  Number var iterates, ivar=0 means none.
Xhomoskedastic      Code homoskedastic or heteroskedastic.
X0                  Number of moving average terms MA for var estimate.
Xnone               Code none or Parzen, none when MA>0 is unwise.
X1.000000e-13       Convergence tolerance, tol=1.0e-8 is reasonable.
X1.000000e-10       Inversion tolerance, eps=1.0e-13 is reasonable
Xminimal            How much output?  Code none, minimal, or full.
X8                  Computed rank of V.
Xuncorrected        Divisor of var, corrected or uncorrected.
X                   Blank line.
X                   Blank line.
X                   Blank line.
X  -2.9245814501616803e+00          -2.92458145  theta(1)
X  -1.2867460434063589e+00          -1.28674604  theta(2)
X   8.1856990357452608e-01           0.81856990  theta(3)
X   3.6115785200404138e-01           0.36115785  theta(4)
X  -1.5375885808695555e+00          -1.53758858  theta(5)
X  -1.0489590970263374e+00          -1.04895910  theta(6)
X   3.0086700241162653e-02           0.03008670  theta(7)
X  -4.6742011797582117e-01          -0.46742012  theta(8)
X   1.6492462994112111e-01           0.16492463  var(1,1)
X   9.2005729031321579e-02           0.09200573  var(2,1)
X   9.2005729031321579e-02           0.09200573  var(1,2)
X   8.9642642316875812e-02           0.08964264  var(2,2)
X   7.6048540298630313e-02           0.07604854  V(1,1)
X  -3.9441553460601381e-02          -0.03944155  V(2,1)
X  -8.1363244658375945e-03          -0.00813632  V(3,1)
X   2.2551454088996347e-03           0.00225515  V(4,1)
X   1.5161071040894071e-02           0.01516107  V(5,1)
X  -8.6386978244942293e-04          -0.00086387  V(6,1)
X  -2.9360670383682128e-03          -0.00293607  V(7,1)
X  -2.1820501983515709e-03          -0.00218205  V(8,1)
X  -3.9441553460601381e-02          -0.03944155  V(1,2)
X   5.0609776285336930e-02           0.05060978  V(2,2)
X   2.1939598380520749e-03           0.00219396  V(3,2)
X   7.1346736728890531e-04           0.00071347  V(4,2)
X  -2.4452170498858623e-03          -0.00244522  V(5,2)
X   1.2263248643774908e-02           0.01226325  V(6,2)
X   3.2176432662395357e-03           0.00321764  V(7,2)
X   1.7266330377801262e-03           0.00172663  V(8,2)
X  -8.1363244658375945e-03          -0.00813632  V(1,3)
X   2.1939598380520749e-03           0.00219396  V(2,3)
X   6.4415535695106418e-03           0.00644155  V(3,3)
X   7.8138028913347761e-04           0.00078138  V(4,3)
X  -6.0353186597764865e-03          -0.00603532  V(5,3)
X  -2.3859940684536310e-03          -0.00238599  V(6,3)
X  -7.5345196703219372e-04          -0.00075345  V(7,3)
X   8.0829924747999706e-04           0.00080830  V(8,3)
X   2.2551454088996347e-03           0.00225515  V(1,4)
X   7.1346736728890531e-04           0.00071347  V(2,4)
X   7.8138028913347761e-04           0.00078138  V(3,4)
X   9.0343959846035443e-04           0.00090344  V(4,4)
X  -4.8087021200371650e-04          -0.00048087  V(5,4)
X   1.0129070210761102e-03           0.00101291  V(6,4)
X  -7.9073548924408663e-04          -0.00079074  V(7,4)
X   2.7549163836087631e-04           0.00027549  V(8,4)
X   1.5161071040894071e-02           0.01516107  V(1,5)
X  -2.4452170498858623e-03          -0.00244522  V(2,5)
X  -6.0353186597764865e-03          -0.00603532  V(3,5)
X  -4.8087021200371650e-04          -0.00048087  V(4,5)
X   8.3213669176229327e-03           0.00832137  V(5,5)
X   2.0084691585193232e-03           0.00200847  V(6,5)
X   1.1068054871987483e-03           0.00110681  V(7,5)
X  -1.1537431897854821e-03          -0.00115374  V(8,5)
X  -8.6386978244942293e-04          -0.00086387  V(1,6)
X   1.2263248643774908e-02           0.01226325  V(2,6)
X  -2.3859940684536310e-03          -0.00238599  V(3,6)
X   1.0129070210761102e-03           0.00101291  V(4,6)
X   2.0084691585193232e-03           0.00200847  V(5,6)
X   6.8944369024945473e-03           0.00689444  V(6,6)
X   1.5596012364267730e-04           0.00015596  V(7,6)
X   4.9917336963915457e-04           0.00049917  V(8,6)
X  -2.9360670383682128e-03          -0.00293607  V(1,7)
X   3.2176432662395357e-03           0.00321764  V(2,7)
X  -7.5345196703219372e-04          -0.00075345  V(3,7)
X  -7.9073548924408663e-04          -0.00079074  V(4,7)
X   1.1068054871987483e-03           0.00110681  V(5,7)
X   1.5596012364267730e-04           0.00015596  V(6,7)
X   1.2861607343363046e-03           0.00128616  V(7,7)
X  -2.4275726609718795e-04          -0.00024276  V(8,7)
X  -2.1820501983515709e-03          -0.00218205  V(1,8)
X   1.7266330377801262e-03           0.00172663  V(2,8)
X   8.0829924747999706e-04           0.00080830  V(3,8)
X   2.7549163836087631e-04           0.00027549  V(4,8)
X  -1.1537431897854821e-03          -0.00115374  V(5,8)
X   4.9917336963915457e-04           0.00049917  V(6,8)
X  -2.4275726609718795e-04          -0.00024276  V(7,8)
X   3.6531994032499421e-04           0.00036532  V(8,8)
X   7.1212458944209244e-09           0.00000001  D(1)
X  -9.5291010557382897e-09          -0.00000001  D(2)
X  -1.5778368693623276e-09          -0.00000000  D(3)
X  -6.4433263552306315e-10          -0.00000000  D(4)
X   1.5722264246975941e-09           0.00000000  D(5)
X  -2.4312439036197547e-09          -0.00000000  D(6)
X  -1.5845329475213192e-10          -0.00000000  D(7)
X  -8.7707819833213637e-10          -0.00000000  D(8)
X   4.4685695288184939e+02         446.85695288  obj
END_OF_FILE
if test 5775 -ne `wc -c <'ch5eg1/ending.bak'`; then
    echo shar: \"'ch5eg1/ending.bak'\" unpacked with wrong size!
fi
# end of 'ch5eg1/ending.bak'
fi
if test -f 'ch6eg1/ending.bak' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'ch6eg1/ending.bak'\"
else
echo shar: Extracting \"'ch6eg1/ending.bak'\" \(5772 characters\)
sed "s/^X//" >'ch6eg1/ending.bak' <<'END_OF_FILE'
XGallant, "Nonlinear Statistical Models," Chapter 6, Figure 1, p. 436.
XTSLS               What estimation method?  Code SUR, TSLS, or GMM.
X220                Number of observations, t = 1, ..., n.
X2                  Number of equations, i.e. dimension of e.
X17                 Number of instruments, i.e. dimension of Z.
X8                  Number of parameters, i.e. dimension of theta.
X20                 Upper limit on Gauss-Newton iterations.
X1                  Number var iterates, ivar=0 means none.
Xhomoskedastic      Code homoskedastic or heteroskedastic.
X0                  Number of moving average terms MA for var estimate.
Xnone               Code none or Parzen, none when MA>0 is unwise.
X1.000000e-08       Convergence tolerance, tol=1.0e-8 is reasonable.
X1.000000e-10       Inversion tolerance, eps=1.0e-13 is reasonable
Xfull               How much output?  Code none, minimal, or full.
X8                  Computed rank of V.
Xuncorrected        Divisor of var, corrected or uncorrected.
X                   Blank line.
X                   Blank line.
X                   Blank line.
X  -2.1378778275860273e+00          -2.13787783  theta(1)
X  -1.9893911556433312e+00          -1.98939116  theta(2)
X   7.0939153916745556e-01           0.70939154  theta(3)
X   3.3663399323270310e-01           0.33663399  theta(4)
X  -1.4019968694481941e+00          -1.40199687  theta(5)
X  -1.1388969627624823e+00          -1.13889696  theta(6)
X   2.9131488559477250e-02           0.02913149  theta(7)
X  -5.0050229673416280e-01          -0.50050230  theta(8)
X   1.6847307354905444e-01           0.16847307  var(1,1)
X   9.4991006557633997e-02           0.09499101  var(2,1)
X   9.4991006557633997e-02           0.09499101  var(1,2)
X   9.3711564655296287e-02           0.09371156  var(2,2)
X   3.4123357521137065e-01           0.34123358  V(1,1)
X  -4.0515026192173231e-01          -0.40515026  V(2,1)
X  -7.4443670981270527e-02          -0.07444367  V(3,1)
X  -1.7075137202523048e-02          -0.01707514  V(4,1)
X   7.3616408974018430e-02           0.07361641  V(5,1)
X  -8.0520944995638929e-02          -0.08052094  V(6,1)
X  -1.3988025621517654e-02          -0.01398803  V(7,1)
X  -2.2077927314748533e-02          -0.02207793  V(8,1)
X  -4.0515026192173231e-01          -0.40515026  V(1,2)
X   5.6591782097119825e-01           0.56591782  V(2,2)
X   9.3113487046268872e-02           0.09311349  V(3,2)
X   2.8410019934882371e-02           0.02841002  V(4,2)
X  -8.0617530766411080e-02          -0.08061753  V(5,2)
X   1.2644653136927941e-01           0.12644653  V(6,2)
X   1.9325013502098160e-02           0.01932501  V(7,2)
X   2.9896132263088925e-02           0.02989613  V(8,2)
X  -7.4443670981270527e-02          -0.07444367  V(1,3)
X   9.3113487046268872e-02           0.09311349  V(2,3)
X   2.4067454001684946e-02           0.02406745  V(3,3)
X   5.7835699332500186e-03           0.00578357  V(4,3)
X  -2.1581815564116969e-02          -0.02158182  V(5,3)
X   1.7403224408106693e-02           0.01740322  V(6,3)
X   1.9150094956931823e-03           0.00191501  V(7,3)
X   5.9076727860237991e-03           0.00590767  V(8,3)
X  -1.7075137202523048e-02          -0.01707514  V(1,4)
X   2.8410019934882371e-02           0.02841002  V(2,4)
X   5.7835699332500186e-03           0.00578357  V(3,4)
X   2.5487432759845568e-03           0.00254874  V(4,4)
X  -4.7950486640943122e-03          -0.00479505  V(5,4)
X   7.3439708547267122e-03           0.00734397  V(6,4)
X  -5.1094810414861448e-05          -0.00005109  V(7,4)
X   1.8674253117946520e-03           0.00186743  V(8,4)
X   7.3616408974018430e-02           0.07361641  V(1,5)
X  -8.0617530766411080e-02          -0.08061753  V(2,5)
X  -2.1581815564116969e-02          -0.02158182  V(3,5)
X  -4.7950486640943122e-03          -0.00479505  V(4,5)
X   2.2762057613803403e-02           0.02276206  V(5,5)
X  -1.5195066143983841e-02          -0.01519507  V(6,5)
X  -9.9415605596577862e-04          -0.00099416  V(7,5)
X  -5.6251801394123823e-03          -0.00562518  V(8,5)
X  -8.0520944995638929e-02          -0.08052094  V(1,6)
X   1.2644653136927941e-01           0.12644653  V(2,6)
X   1.7403224408106693e-02           0.01740322  V(3,6)
X   7.3439708547267122e-03           0.00734397  V(4,6)
X  -1.5195066143983841e-02          -0.01519507  V(5,6)
X   3.3346276344180567e-02           0.03334628  V(6,6)
X   3.7331467729901690e-03           0.00373315  V(7,6)
X   6.9269333322694495e-03           0.00692693  V(8,6)
X  -1.3988025621517654e-02          -0.01398803  V(1,7)
X   1.9325013502098160e-02           0.01932501  V(2,7)
X   1.9150094956931823e-03           0.00191501  V(3,7)
X  -5.1094810414861448e-05          -0.00005109  V(4,7)
X  -9.9415605596577862e-04          -0.00099416  V(5,7)
X   3.7331467729901690e-03           0.00373315  V(6,7)
X   2.0417788436814010e-03           0.00204178  V(7,7)
X   6.0905830809060847e-04           0.00060906  V(8,7)
X  -2.2077927314748533e-02          -0.02207793  V(1,8)
X   2.9896132263088925e-02           0.02989613  V(2,8)
X   5.9076727860237991e-03           0.00590767  V(3,8)
X   1.8674253117946520e-03           0.00186743  V(4,8)
X  -5.6251801394123823e-03          -0.00562518  V(5,8)
X   6.9269333322694495e-03           0.00692693  V(6,8)
X   6.0905830809060847e-04           0.00060906  V(7,8)
X   2.0036106220501352e-03           0.00200361  V(8,8)
X   1.6787206308357473e-07           0.00000017  D(1)
X  -2.4625529712808257e-07          -0.00000025  D(2)
X  -4.6504414197150224e-08          -0.00000005  D(3)
X  -1.4610224422905096e-08          -0.00000001  D(4)
X   4.0400010629413197e-08           0.00000004  D(5)
X  -5.8821932254927893e-08          -0.00000006  D(6)
X  -8.2632941559150507e-09          -0.00000001  D(7)
X  -1.6640001385280539e-08          -0.00000002  D(8)
X   3.5611514196184793e+01          35.61151420  obj
END_OF_FILE
if test 5772 -ne `wc -c <'ch6eg1/ending.bak'`; then
    echo shar: \"'ch6eg1/ending.bak'\" unpacked with wrong size!
fi
# end of 'ch6eg1/ending.bak'
fi
if test -f 'ch6eg2/detail3l.bak' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'ch6eg2/detail3l.bak'\"
else
echo shar: Extracting \"'ch6eg2/detail3l.bak'\" \(7043 characters\)
sed "s/^X//" >'ch6eg2/detail3l.bak' <<'END_OF_FILE'
X
X
X     **********************************************************************
X     *                                                                    *
X     *                             nlmdl 2.0                              *
X     *                                                                    *
X     **********************************************************************
X
X
X     **********************************************************************
X     *                                                                    *
X     *                         Parameter settings                         *
X     *                                                                    *
X     **********************************************************************
X
X     Gallant, "Nonlinear Statistical Models," Chapter 6, Figure 2, p. 449-450.
X     GMM                What estimation method?  Code SUR, TSLS, or GMM.
X     238                Number of observations, t = 1, ..., n.
X     1                  Number of equations, i.e. dimension of e.
X     3                  Number of instruments, i.e. dimension of Z.
X     2                  Number of parameters, i.e. dimension of theta.
X     20                 Upper limit on Gauss-Newton iterations.
X     1                  Number var iterates, ivar=0 means none.
X     heteroskedastic    Code homoskedastic or heteroskedastic.
X     3                  Number of moving average terms MA for var estimate.
X     Parzen             Code none or Parzen, none when MA>0 is unwise.
X     1.000000e-05       Convergence tolerance, tol=1.0e-8 is reasonable.
X     1.000000e-13       Inversion tolerance, eps=1.0e-13 is reasonable
X     full               How much output?  Code none, minimal, or full.
X
X
X     **********************************************************************
X     *                                                                    *
X     *                           Starting theta                           *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                           Col  1
X 
X                               Row  1   -0.400000
X                               Row  2    0.900000
X
Xvar_loop 0
X   2.3800000000000000e+02         238.00000000  var(1,1)
X   2.3848513503508150e+02         238.48513504  var(2,1)
X   2.3866807009536583e+02         238.66807010  var(3,1)
X   2.3848513503508150e+02         238.48513504  var(1,2)
X   2.3897620940526255e+02         238.97620941  var(2,2)
X   2.3916481122626234e+02         239.16481123  var(3,2)
X   2.3866807009536583e+02         238.66807010  var(1,3)
X   2.3916481122626234e+02         239.16481123  var(2,3)
X   2.3975964092305995e+02         239.75964092  var(3,3)
X
Xtheta_loop 0
X  -4.0000000000000002e-01          -0.40000000  theta(1)
X   9.0000000000000002e-01           0.90000000  theta(2)
X   2.2977593855825589e+00           2.29775939  obj
X  -4.9910043429446205e-01          -0.49910043  D(1)
X   9.8933364963875864e-02           0.09893336  D(2)
XStep length = 1 
X
Xtheta_loop 1
X  -8.9910043429446207e-01          -0.89910043  theta(1)
X   9.9893336496387586e-01           0.99893336  theta(2)
X   2.7751760221041621e-03           0.00277518  obj
X   5.0366043831498272e-02           0.05036604  D(1)
X  -4.4414610501568677e-06          -0.00000444  D(2)
XStep length = 1 
X
Xtheta_loop 2
X  -8.4873439046296384e-01          -0.84873439  theta(1)
X   9.9892892350282569e-01           0.99892892  theta(2)
X   2.7717868638761308e-03           0.00277179  obj
X  -1.1734797022229638e-04          -0.00011735  D(1)
X   2.1096605255709332e-07           0.00000021  D(2)
XStep length = 1 
X
Xtheta_loop 3
X  -8.4885173843318618e-01          -0.84885174  theta(1)
X   9.9892913446887821e-01           0.99892913  theta(2)
X   2.7717868580610532e-03           0.00277179  obj
X   2.6798358348564710e-07           0.00000027  D(1)
X  -5.5350439549602871e-10          -0.00000000  D(2)
XTolerence check passed.
X
Xvar_loop 1
X   4.3978220854904504e-01           0.43978221  var(1,1)
X   4.4050039895600307e-01           0.44050040  var(2,1)
X   4.3420956263910121e-01           0.43420956  var(3,1)
X   4.4050039895600307e-01           0.44050040  var(1,2)
X   4.4122700259798669e-01           0.44122700  var(2,2)
X   4.3493663977747488e-01           0.43493664  var(3,2)
X   4.3420956263910121e-01           0.43420956  var(1,3)
X   4.3493663977747488e-01           0.43493664  var(2,3)
X   4.2967081964723475e-01           0.42967082  var(3,3)
X
Xtheta_loop 0
X  -8.4885173843318618e-01          -0.84885174  theta(1)
X   9.9892913446887821e-01           0.99892913  theta(2)
X   1.2014636301992196e+00           1.20146363  obj
X  -6.4020486065461402e-03          -0.00640205  D(1)
X  -8.8353145570584189e-04          -0.00088353  D(2)
XStep length = 1 
X
Xtheta_loop 1
X  -8.5525378703973232e-01          -0.85525379  theta(1)
X   9.9804560301317236e-01           0.99804560  theta(2)
X   1.0861341198092420e+00           1.08613412  obj
X   2.6747016119385693e-06           0.00000267  D(1)
X  -1.7073275785211286e-08          -0.00000002  D(2)
XTolerence check passed.
X
X
X     **********************************************************************
X     *                                                                    *
X     *                               theta                                *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                           Col  1
X 
X                               Row  1   -0.855254
X                               Row  2    0.998046
X
X
X
X     **********************************************************************
X     *                                                                    *
X     *                                var                                 *
X     *                (no degrees of freedom corrections)                 *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                               Col  1      Col  2      Col  3
X 
X                   Row  1    0.439782    0.440500    0.434210
X                   Row  2    0.440500    0.441227    0.434937
X                   Row  3    0.434210    0.434937    0.429671
X
X
X
X     **********************************************************************
X     *                                                                    *
X     *                                 V                                  *
X     *                             (rank = 2)                             *
X     *                                                                    *
X     **********************************************************************
X 
X 
X                                     Col  1      Col  2
X 
X                         Row  1     3.60663 -0.00715768
X                         Row  2 -0.00715768   2.117e-05
X
END_OF_FILE
if test 7043 -ne `wc -c <'ch6eg2/detail3l.bak'`; then
    echo shar: \"'ch6eg2/detail3l.bak'\" unpacked with wrong size!
fi
# end of 'ch6eg2/detail3l.bak'
fi
if test -f 'display.cc' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'display.cc'\"
else
echo shar: Extracting \"'display.cc'\" \(4567 characters\)
sed "s/^X//" >'display.cc' <<'END_OF_FILE'
X/* ----------------------------------------------------------------------------
X
Xnlmdl: display.cc
X
Xnlmdl is a C++ implementation of the statistical methods in A. Ronald 
XGallant, "Nonlinear Statistical Models," New York: John Wiley and Sons, 
X1987, ISBN 0-471-80260-3, using a matrix class realmat that is distributed 
Xwith it.  The header files nlmdl.h and realmat.h describe the use of the 
Xprogram and matrix class, respectively.  
X
XCopyright (C) 1990.
X
XA. Ronald Gallant
XP.O. Box 5513 
XRaleigh NC 27650-5513 
XUSA   
X
XPermission to use, copy, modify, and distribute this software and its 
Xdocumentation for any purpose and without fee is hereby granted, provided 
Xthat the above copyright notice appear in all copies and that both that 
Xcopyright notice and this permission notice appear in supporting 
Xdocumentation.  
X
XThis software is provided "as is" without any expressed or implied warranty.
X
X---------------------------------------------------------------------------- */
X/* display.cc is a piece of status.cc that was split */
X/* out to reduce Turbo C++ memory requirements.      */
X
X#include "status.h"
X
Xint status::display(display_mode mode)
X{
X
X  INTEGER i,j,len;
X  char    temp[MAX_STATUS_LINE];
X  char*   pad = "     ";
X
X
X  switch (mode) {
X
X  case START_UP:
X
X    cout << starbox("/Parameter settings//_") << "\n";
X
X    cout<<pad<< switches << "\n";
X
X    sprintf(temp,"%-19s", method);
X    cout<<pad<<temp<< "What estimation method?  Code SUR, TSLS, or GMM.\n";
X
X    sprintf(temp,"%-19d", n);
X    cout<<pad<<temp<< "Number of observations, t = 1, ..., n.\n";
X
X    sprintf(temp,"%-19d",M);
X    cout<<pad<<temp<< "Number of equations, i.e. dimension of e.\n";
X
X    sprintf(temp,"%-19d",K);
X    cout<<pad<<temp<<"Number of instruments, i.e. dimension of Z.\n";
X
X    sprintf(temp,"%-19d",p);
X    cout<<pad<<temp<<"Number of parameters, i.e. dimension of theta.\n";
X
X    sprintf(temp,"%-19d",itheta);
X    cout<<pad<<temp<< "Upper limit on Gauss-Newton iterations.\n";
X
X    sprintf(temp,"%-19d",ivar);
X    cout<<pad<<temp<< "Number var iterates, ivar=0 means none.\n";
X
X    sprintf(temp,"%-19s",vartype);
X    cout<<pad<<temp<< "Code homoskedastic or heteroskedastic.\n";
X
X    sprintf(temp,"%-19d",MA);
X    cout<<pad<<temp<< "Number of moving average terms MA for var estimate.\n";
X
X    sprintf(temp,"%-19s",weights);
X    cout<<pad<<temp<< "Code none or Parzen, none when MA>0 is unwise.\n";
X
X    sprintf(temp,"%-19.6e",tol);
X    cout<<pad<<temp<< "Convergence tolerance, tol=1.0e-8 is reasonable.\n";
X
X    sprintf(temp,"%-19.6e",eps);
X    cout<<pad<<temp<< "Inversion tolerance, eps=1.0e-13 is reasonable\n";
X
X    sprintf(temp,"%-19s",detail);
X    cout<<pad<<temp<< "How much output?  Code none, minimal, or full.\n";
X
X    break;
X
X  case VAR_ITERATE:
X
X    if ( rows(var) > 0 && cols(var) > 0 ){
X      if ( strcmp(method,"SUR") == 0 || strcmp(method,"TSLS") == 0){
X        for (j=1; j<=M; j++){ 
X        for (i=1; i<=M; i++){ 
X          sprintf(temp,"%25.16e % 20.8f ",var.check2(i,j),var.elem(i,j));
X          cout<<temp<< " var(" << i << "," << j << ")\n";
X        }
X        }  
X      }
X      else{
X        len = M*K;
X        for (j=1; j<=len; j++){
X        for (i=1; i<=len; i++){
X          sprintf(temp,"%25.16e % 20.8f ",var.check2(i,j),var.elem(i,j));
X          cout<<temp<< " var(" << i << "," << j << ")\n";
X        }
X        }
X      }
X    }
X    else{
X      cout << "Error, status::display, var is not initialized.\n";
X    }
X
X    break;
X
X   case THETA_ITERATE:
X
X    if ( rows(theta) > 0 && cols(theta) > 0 ){
X      for (i=1; i<=p; i++){
X        sprintf(temp,"%25.16e % 20.8f ",theta[i],theta[i]);
X        cout<<temp<< " theta(" << i << ")\n";
X      }
X    }
X    else{
X      cout << "Error, status::display, theta is not initialized.\n";
X    }
X
X    sprintf(temp,"%25.16e % 20.8f ",obj,obj);
X    cout<<temp<< " obj\n";
X
X    if (strcmp(detail,"full")!=0) break;
X
X    if ( rows(D) > 0 && cols(D) > 0 ){
X      for (i=1; i<=p; i++){
X        sprintf(temp,"%25.16e % 20.8f ",D[i],D[i]);
X        cout<<temp<< " D(" << i << ")\n";
X      }
X    }
X    else{
X      cout << "Error, status::display, D is not initialized.\n";
X    }
X 
X    break;
X
X  case TERMINATION:
X  
X    cout << starbox("/theta//_") << theta << "\n";
X
X    if (strcmp(df,"corrected") == 0)
X      cout << starbox("/var/(corrected for degrees of freedom)//_");
X    else
X      cout << starbox("/var/(no degrees of freedom corrections)//_"); 
X
X    cout << var << "\n";
X
X    sprintf(temp,"/V/(rank = %d)//_",rank);
X    cout << starbox(temp) << V << "\n";
X
X    break;
X
X  }
X
X  cout.flush();
X
X  return 0;
X
X}
END_OF_FILE
if test 4567 -ne `wc -c <'display.cc'`; then
    echo shar: \"'display.cc'\" unpacked with wrong size!
fi
# end of 'display.cc'
fi
if test -f 'nlmdl.h' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'nlmdl.h'\"
else
echo shar: Extracting \"'nlmdl.h'\" \(5856 characters\)
sed "s/^X//" >'nlmdl.h' <<'END_OF_FILE'
X/* ---------------------------------------------------------------------------
X
Xnlmdl: nlmdl.h
X
Xnlmdl is a C++ implementation of the statistical methods in A. Ronald 
XGallant, "Nonlinear Statistical Models," New York: John Wiley and Sons, 
X1987, ISBN 0-471-80260-3, using a matrix class realmat that is distributed 
Xwith it.  The header files nlmdl.h and realmat.h describe the use of the 
Xprogram and matrix class, respectively.  
X
XCopyright (C) 1990.
X
XA. Ronald Gallant
XP.O. Box 5513 
XRaleigh NC 27650-5513 
XUSA   
X
XPermission to use, copy, modify, and distribute this software and its 
Xdocumentation for any purpose and without fee is hereby granted, provided 
Xthat the above copyright notice appear in all copies and that both that 
Xcopyright notice and this permission notice appear in supporting 
Xdocumentation.  
X
XThis software is provided "as is" without any expressed or implied warranty.
X
X----------------------------------------------------------------------------- 
X
XThis header describes the program nlmdl.cc for estimating theta of the model 
Xe=q(t,theta).  The user supples a file named starting.dat whose contents are 
Xdescribed by the template:
X
X  switches This line is passed to class model as a string.
X  method   What estimation method?  Code SUR, TSLS, or GMM.
X  n        Number of observations, t = 1, ..., n.
X  M        Number of equations in the system, i.e. dimension of e.
X  K        Number of instruments, i.e. dimension of Z, code K=0 with SUR.
X  p        Number of parameters, i.e. dimension of theta.
X  itheta   Upper limit on Gauss-Newton iterations.
X  ivar     Number of var iterates, ivar=0 means none.
X  vartype  Code homoskedastic or heteroskedastic.
X  MA       Number of moving average terms for var estimate, code MA=0 if e iid.
X  weights  Code none or Parzen, coding none when MA>0 is asking for trouble.
X  tol      Convergence tolerance, tol=1.0e-8 is reasonable.
X  eps      Inversion tolerance, eps=1.0e-13 is reasonable (if REAL is double).
X  detail   How much output?  Code none, minimal, or full.
X           Blank line.
X           Blank line.
X           Blank line.
X           Blank line.
X           Blank line.
X  theta(1) Starting values for theta.  These must be supplied.
X           
X    to
X
X  theta(p) 
X  var(1)   Starting values for variance estimate.  These can be omitted.  
X           Matrix is stored columnwise.  If method = SUR or TSLS then then 
X    to     l=M and var corresponds to C(e,e').  If method = GMM then l=M*K 
X           and var corresponds to the variance of sum on t of e Kronecker 
X  var(l*l) product Z.  
X
X
XSee class status for more detail, especially status::from.  The file 
Xstarting.dat is read twice.  At the first reading it must contain at least 
Xthe first line, switches.  Class model, described next, can supply entries 
Xfor the second reading.  
X
XThe user also supplies the class model as files model.h and model.cc.  The 
Xclass model is declared in model.h which should match the following template:
X
X   #include "status.h"
X   external status s;
X
X   class model
X   {   
X   protected:
X     //...
X
X   public:
X            model();          
X            ~model();           
X   realmat  e_t;               // e_t    is s.M by 1
X   realmat  dele_t;            // dele_t is s.M by s.p
X   realmat  Z_t;               // Z_t    is s.K by 1, null if s.K = 0;
X   int      initialize();
X   int      terminate();
X   void     e(INTEGER t);      // e    fills in e_t
X   void     dele(INTEGER t);   // dele fills in dele_t = del e_t wrt theta
X   void     Z(INTEGER t);      // Z    fills in Z_t
X   }
X
XFor SUR, code the function Z as follows:
X
X   realmat model::Z(INTEGER t) { }
X
XIn writing the class model, the relevant facts regarding program flow in 
Xnlmdl.cc are as follows: 
X
X  1. First, s is tentatively filled in by reading starting.dat using 
X
X       status s;
X       s.from(s.starting);
X
X     The primary purpose of this read is to get the first line from 
X     starting.dat so as to make it available to model as the string 
X     s.switches.  Nonetheless, everything in starting.dat is read and 
X     put in s.  
X
X  2. Next, initialize() of class model is called.  The function initialize() 
X     can read starting.dat, can read other data, etc.  Since all of class 
X     status's data is public, any of it can be filled in or changed by 
X     initialize().  One could, for example, fill in all of class status's 
X     data and write it using 
X
X       s.to(s.starting);
X
X     One could also switch to a different starting status file using 
X
X       s.starting="filename";
X
X     or a different ending status file using 
X
X       s.ending="filename";
X
X     If initialize() returns 0, execution stops.
X
X  3. Then, class status's data is definitively filled in by the call
X
X       s.from(s.starting);
X
X     and m.e_t, m.dele_t, and m.Z_t are dimensioned as
X
X       m.e_t.resize(s.M,1);
X       m.dele_t.resize(s.M,s.p);
X       if (s.K > 0) m.Z_t.resize(s.K,1);
X                          
X     and estimation proceeds using these settings and dimensions.  
X
X  4. Finally, terminate() of class model is called; terminate() can read data, 
X     write data, etc.  If terminate() returns 0 execution stops, otherwise 
X     initialize() is called again and the cycle repeats (from Step 2).  One 
X     can use this feature to loop over a grid of starting values.
X
X
XReference:
X
XGallant, A. Ronald (1987), "Nonlinear Statistical Models,"  New York: John 
XWiley and Sons. ISBN 0-471-80260-3.
X
X--------------------------------------------------------------------------- */
X
X
X#ifndef __FILE_NLMDL_H_SEEN__
X#pragma once
X#define __FILE_NLMDL_H_SEEN__ 1
X
X#include "usual.h"      //This header is also in status.h
X#include "realmat.h"    //This header is also in status.h
X#include "tools.h"      //This header is also in status.h
X
X#include "status.h"
X#include "model.h"
X
X#define  VERSION "2.0"
X
X#endif
X
END_OF_FILE
if test 5856 -ne `wc -c <'nlmdl.h'`; then
    echo shar: \"'nlmdl.h'\" unpacked with wrong size!
fi
# end of 'nlmdl.h'
fi
if test -f 'realmat.h' -a "${1}" != "-c" ; then 
  echo shar: Will not clobber existing file \"'realmat.h'\"
else
echo shar: Extracting \"'realmat.h'\" \(5870 characters\)
sed "s/^X//" >'realmat.h' <<'END_OF_FILE'
X/* ----------------------------------------------------------------------------
X
Xrealmat: realmat.h
X
Xrealmat is a C++ matrix class.  The header file realmat.h describes its use.
X
XCopyright (C) 1990.
X
XA. Ronald Gallant
XP.O. Box 5513 
XRaleigh NC 27650-5513 
XUSA   
X
XPermission to use, copy, modify, and distribute this software and its 
Xdocumentation for any purpose and without fee is hereby granted, provided 
Xthat the above copyright notice appear in all copies and that both that 
Xcopyright notice and this permission notice appear in supporting 
Xdocumentation.  
X
XThis software is provided "as is" without any expressed or implied warranty.
X
X------------------------------------------------------------------------------- 
X
XThis header defines a matrix class.  A matrix X with r rows, and c columns is 
Xconstructed using realmat X(r,c).  X is stored columnwise with no wasted space 
Xwhich is to say that vec(X) is what is stored.  Indexing of the elements starts 
Xwith 1, not zero.  The protected members of realmat are rows, cols, len, and x.
X
XAs an illustration, suppose X is constructed using realmat X(r,c).  Then 
XX.rows=r, X.cols=c, X.len=r*c, the pointer X.x points to the first element of 
Xvec(X), X[i] returns the i-th element of vec(X), i=1,...,X.len, X.elem(i,j) 
Xreturns the (i,j)-th element of X, i=1,...,X.rows, j=1,...,.X.cols.  An element 
Xof X is of type REAL and X.rows, X.cols, and X.len are of type INTEGER.  The 
Xtypedef of REAL and INTEGER is in usual.h.  To allocate store at run time, use
Xrealmat X; X.resize(r,c);.
X
XThe operators +, -, *, =, +=, -=, ++, --, ==, !=, <, <=, >, >= and << are 
Xdefined.  T(X) transposes X.  X.check1(i) and X.check2(i,j) access the elements 
Xof vec(X) and X respectively with range checking.  invpsd(X) or invpsd(X,eps) 
Xreturns a g-inverse of a positive semi-definite, symmetric matrix X with 
Xminimal error checking; eps is the tolerance used in rank determination and 
Xdefaults to 1.e-13.  rows(X) returns X.rows and cols(X) returns X.cols.  The 
Xoperator ++ acts like ++X, not like X++; similarly for --.  The operators ==,
X<, <=, >, and >= evaluate to true if the matrices compared are the same size 
Xand the relation is true for each element; (A!=B) = !(A==B).
X
X-----------------------------------------------------------------------------*/
X
X#ifndef __FILE_REALMAT_H_SEEN__
X#pragma once
X#define __FILE_REALMAT_H_SEEN__ 1
X
X#include "usual.h"
X#include "tools.h"
X
Xclass realmat
X{
Xprotected:
X
X  INTEGER       rows;
X  INTEGER       cols;
X  INTEGER       len;
X  REAL          *x;
X
X                realmat(INTEGER r, INTEGER c, REAL* a);
X  void          resize(INTEGER r, INTEGER c, REAL* a); //expects this.x == 0
X
X
Xpublic:
X                realmat();
X
X                realmat(INTEGER r, INTEGER c);
X                realmat(INTEGER r, INTEGER c, REAL fill_value);
X
X                realmat(realmat&);
X
X                ~realmat();
X
Xvoid            resize(INTEGER r, INTEGER c);
Xvoid            resize(INTEGER r, INTEGER c, REAL fill_value);
X
Xrealmat&        operator=(realmat& a);
Xrealmat&        operator+=(realmat& a);
Xrealmat&        operator-=(realmat& a);
Xrealmat&        operator++();
Xrealmat&        operator--();
X
XREAL&           operator[](INTEGER i);
XREAL&           elem(INTEGER i, INTEGER j);
XREAL&           check1(INTEGER i);
XREAL&           check2(INTEGER i, INTEGER j);
X
Xvoid            error(const char* msg);
X
Xfriend int      operator==(realmat& a, realmat& b);
Xfriend int      operator!=(realmat& a, realmat& b);
X
Xfriend int      operator<(realmat& a, realmat& b);
Xfriend int      operator>(realmat& a, realmat& b);
X
Xfriend int      operator<=(realmat& a, realmat& b);
Xfriend int      operator>=(realmat& a, realmat& b);
X
Xfriend INTEGER  rows(realmat& a);
Xfriend INTEGER  cols(realmat& a);
X
Xfriend ostream& operator<<(ostream& stream, realmat& a);
X
Xfriend realmat  operator+(realmat& a, realmat& b);
Xfriend realmat  operator+(realmat& a);
X
Xfriend realmat  operator-(realmat& a, realmat& b);
Xfriend realmat  operator-(realmat& a);
X
Xfriend realmat  operator*(realmat& a, realmat& b);
Xfriend realmat  operator*(REAL&    a, realmat& b);
Xfriend realmat  operator*(INTEGER& a, realmat& b);
X
Xfriend realmat  T(realmat& a);
X
Xfriend realmat  invpsd(realmat& a, REAL eps = 1.0e-13); //a is psd, symmetric
X
X};
X
X
X//protected:
X
Xinline realmat::realmat(INTEGER r, INTEGER c, REAL* a)
X  {rows=r; cols=c; len=r*c; x = a;}
X
Xinline void realmat::resize(INTEGER r, INTEGER c, REAL* a)
X  {rows=r; cols=c; len=r*c; x = a;} //Note: No delete, this.x == 0 expected.
X
X
X//public:
X
Xinline int operator!=(realmat& a, realmat& b)
X  {return !(a == b);}
X
Xinline REAL& realmat::operator[](INTEGER i)
X  {return x[i-1];}
X
Xinline REAL& realmat::elem(INTEGER i, INTEGER j)
X  {return x[i + rows*j - rows - 1];}   // returns x[rows*(j-1)+i-1]
X
Xinline INTEGER rows(realmat& a)
X  {return a.rows;}
X
Xinline INTEGER cols(realmat& a)
X  {return a.cols;}
X
Xinline realmat::~realmat()
X  {delete x;} 
X
Xinline realmat::realmat()
X  {rows=0; cols=0; len=0; x = 0;}
X
Xinline realmat::realmat(INTEGER r, INTEGER c) 
X{
X  if (r<=0) error("Error, realmat, realmat, Number of rows not positive");
X  if (c<=0) error("Error, realmat, realmat, Number of columns not positive");
X  rows=r; 
X  cols=c; 
X  len=r*c; 
X  x = new REAL[len];
X  if (x == 0) error("Error, realmat, realmat, Operator new failed");
X}
X
Xinline void realmat::resize(INTEGER r, INTEGER c)
X{
X  if (r<=0) error("Error, realmat, resize, Number of rows not positive");
X  if (c<=0) error("Error, realmat, resize, Number of columns not positive");
X  delete x;
X  rows=r; 
X  cols=c; 
X  len=r*c; 
X  x = new REAL[len];
X  if (x == 0) error("Error, realmat, resize, Operator new failed");
X}
X
Xextern void default_realmat_error_handler(const char* msg);
Xextern ONE_ARG_ERROR_HANDLER_T realmat_error_handler;
Xextern ONE_ARG_ERROR_HANDLER_T 
X  set_realmat_error_handler(ONE_ARG_ERROR_HANDLER_T f);
X
X#endif
END_OF_FILE
if test 5870 -ne `wc -c <'realmat.h'`; then
    echo shar: \"'realmat.h'\" unpacked with wrong size!
fi
# end of 'realmat.h'
fi
echo shar: End of archive 2 \(of 6\).
cp /dev/null ark2isdone
MISSING=""
for I in 1 2 3 4 5 6 ; do
    if test ! -f ark${I}isdone ; then
	MISSING="${MISSING} ${I}"
    fi
done
if test "${MISSING}" = "" ; then
    echo You have unpacked all 6 archives.
    rm -f ark[1-9]isdone
else
    echo You still need to unpack the following archives:
    echo "        " ${MISSING}
fi
##  End of shell archive.
exit 0

exit 0 # Just in case...
-- 
Kent Landfield                   INTERNET: kent at sparky.IMD.Sterling.COM
Sterling Software, IMD           UUCP:     uunet!sparky!kent
Phone:    (402) 291-8300         FAX:      (402) 291-4362
Please send comp.sources.misc-related mail to kent at uunet.uu.net.



More information about the Comp.sources.misc mailing list